Background and Data

COVID-19 is causing havoc in Oregon once again, and as numbers continue to spike, I decided to revisit one the of the first projects I worked on in R Studio. That project can be seen here, and used data from Johns Hopkins. However, because that data is no longer updated this investigation will use data from the NY times that has more current data. The repository for the NY Times data can be found here, and the datasets that are being included are :

  1. us.states : state level data (file description here)

  2. us.counties : county-level data (file description here)

  3. colleges : number of reported cases among students and employees at American colleges and universities, updated May 26th (file description here)

  4. mask_use : survey between July 2 and July 14 (2020) where participants were asked, “How oftern do you wear a mask in public when you expect to be within six feet of another person?” (file description here)

  5. vacc: state level COVID-19 daily vaccination numbers time series data from the Johns Hopkins University repository (file description here)

  6. policytrackerOR: dates and description of policies going into/out of effect in Oregon. To load data for a particular state go to here, find the name of the state file you want to work with.

Here is the data :

us.states <- read_csv('https://raw.githubusercontent.com/nytimes/covid-19-data/master/us-states.csv')
us.counties <- read_csv('https://raw.githubusercontent.com/nytimes/covid-19-data/master/us-counties.csv')
colleges <- read_csv('https://raw.githubusercontent.com/nytimes/covid-19-data/master/colleges/colleges.csv')
mask_use <- read_csv('https://raw.githubusercontent.com/nytimes/covid-19-data/master/colleges/colleges.csv')
vacc <- read_csv("https://raw.githubusercontent.com/govex/COVID-19/master/data_tables/vaccine_data/us_data/time_series/people_vaccinated_us_timeline.csv")
policytrackerOR <- read_csv("https://raw.githubusercontent.com/govex/COVID-19/govex_data/data_tables/policy_data/table_data/Current/Oregon_policy.csv")

The projected cited above mainly looked at the case, death, and vaccination numbers per state to compare highly and mildly impacted states. In this project I will look at highly impacted states and the counties of Oregon. Additionally this project will use population and density data from the tidycensus package. I discuss more about how I got this data using an API in a blog post here. Note that this data is from 2019, which is a couple years older than the COVID data.

# tidycensus
# State : POP and DENSITY data 
state.pop <- get_estimates(geography = "state", year = 2019, variable =  "POP") %>% rename ("state" = NAME, "population" = value)
state.den <- get_estimates(geography = "state", year = 2019, variable =  "DENSITY") %>% rename ("state" = NAME, "density" = value)
# OREGON : POP and Density data 
or.county.pop <- get_estimates(geography = "county", state = "OR", year = 2019, variable = "POP") %>% rename ("county" = NAME, "population" = value)
or.county.den <- get_estimates(geography = "county", state = "OR", year = 2019, variable = "DENSITY") %>% rename ("county" = NAME, "density" = value)

Wrangling the Data

The COVID data set is already in long form (meaning the dates are in rows instead of columns), and the date is already saved as a date variable. Therefore the main tasks here are to join the us.states data set with the vaccination records, and population estimates. Then join the us.states.vacc with the population data and create new percentage columns.

## # A tibble: 5 × 12
##   date       state      population density cases     case.per deaths perc.deaths
##   <date>     <chr>           <dbl>   <dbl> <dbl>        <dbl>  <dbl>       <dbl>
## 1 2020-01-21 Washington    7614893    115.     1 0.000000131       0           0
## 2 2020-01-22 Washington    7614893    115.     1 0.000000131       0           0
## 3 2020-01-23 Washington    7614893    115.     1 0.000000131       0           0
## 4 2020-01-24 Illinois     12671821    228.     1 0.0000000789      0           0
## 5 2020-01-24 Washington    7614893    115.     1 0.000000131       0           0
## # … with 4 more variables: full.vacc <dbl>, full.vacc.perc <dbl>,
## #   part.vacc <dbl>, part.vacc.perc <dbl>

Note: People_Fully_Vaccinated and People_Partially_Vaccinated show as NA because of the dates displayed were before the vaccine was released.

Next, to join the Oregon county population data with the density data, remove “County, Oregon” from each county object, filter out Oregon from the us.counties data, and then join with the population data per county.

## # A tibble: 10 × 8
##    date       county     population density cases cases.perc deaths deaths.perc
##    <date>     <chr>           <dbl>   <dbl> <dbl>      <dbl>  <dbl>       <dbl>
##  1 2020-02-28 Washington     601592   831.      1 0.00000166      0           0
##  2 2020-02-29 Washington     601592   831.      1 0.00000166      0           0
##  3 2020-03-01 Washington     601592   831.      2 0.00000332      0           0
##  4 2020-03-02 Washington     601592   831.      2 0.00000332      0           0
##  5 2020-03-03 Washington     601592   831.      2 0.00000332      0           0
##  6 2020-03-04 Washington     601592   831.      2 0.00000332      0           0
##  7 2020-03-05 Washington     601592   831.      2 0.00000332      0           0
##  8 2020-03-06 Washington     601592   831.      2 0.00000332      0           0
##  9 2020-03-07 Jackson        220944    79.4     2 0.00000905      0           0
## 10 2020-03-07 Klamath         68238    11.5     1 0.0000147       0           0

Using the data

Looking at States

To begin lets look at the country as a whole, by state. The data will be filtered for 2022-03-07, and then lets look at the top five states with :

Most number of cases

Most Cases
By State as of 2022-03-07
State Total Population Cases Percentage
California 39,512,223 9,013,699 22.81%
Texas 28,995,881 6,595,988 22.75%
Florida 21,477,737 5,818,732 27.09%
New York 19,453,561 4,921,272 25.30%
Illinois 12,671,821 3,045,351 24.03%

Highest percentage of cases

Highest Percent of Cases
By State as of 2022-03-07
State Total Population Cases Percentage
Rhode Island 1,059,361 357,136 33.71%
Alaska 731,545 239,705 32.77%
North Dakota 762,062 238,784 31.33%
Kentucky 4,467,673 1,299,116 29.08%
Utah 3,205,958 924,978 28.85%

Most number of deaths

Most Deaths
By State as of 2022-03-07
State Total Population Deaths Percentage
California 39,512,223 86,731 0.22%
Texas 28,995,881 85,878 0.30%
Florida 21,477,737 71,326 0.33%
New York 19,453,561 66,955 0.34%
Pennsylvania 12,801,989 43,600 0.34%

Highest percentage of deaths

Highest percent of Deaths
By State as of 2022-03-07
State Total Population Deaths Percentage
Mississippi 2,976,149 12,171 0.41%
Arizona 7,278,717 27,947 0.38%
Alabama 4,903,185 18,569 0.38%
New Jersey 8,882,190 33,027 0.37%
Louisiana 4,648,794 16,785 0.36%

Most people fully vaccinated

Most People Vaccinated
By State as of 2022-03-07
State Total Population People Fully Vaccinated Percentage
California 39,512,223 27,915,579 70.65%
Texas 28,995,881 17,472,130 60.26%
New York 19,453,561 14,730,525 75.72%
Florida 21,477,737 14,194,505 66.09%
Pennsylvania 12,801,989 8,596,920 67.15%

Highest percentage of population fully vaccinated

Highest Percent of People Fully Vaccinated
By State as of 2022-03-07
State Total Population People Fully Vaccinated Percentage
District of Columbia 705,749 643,121 91.13%
Puerto Rico 3,193,694 2,636,414 82.55%
Rhode Island 1,059,361 857,050 80.90%
Vermont 623,989 501,425 80.36%
Maine 1,344,212 1,054,909 78.48%

Looking at Oregon Counties

Next to look at the data a little closer to home, for Oregon Counties. Initially filtering by the most recent date, which as of this being written is 2022-03-07`, looking at a graph of the state as a whole, and then look at the top five Oregon counties.

Most number of cases

Most Cases
Oregon Counties as of 2022-03-07
County Total Population Cases Percentage
Multnomah 812,855 111,440 13.71%
Washington 601,592 84,510 14.05%
Marion 347,818 68,454 19.68%
Clackamas 418,187 59,381 14.20%
Lane 382,067 56,538 14.80%

Highest percentage of Cases

Highest Percentage of Cases
Oregon Counties as of 2022-03-07
County Total Population Cases Percentage
Jefferson 24,658 7,161 29.04%
Umatilla 77,950 22,256 28.55%
Malheur 30,571 8,181 26.76%
Morrow 11,603 2,943 25.36%
Crook 24,404 6,001 24.59%

Most number of Deaths

Most Deaths
Oregon Counties as of 2022-03-07
County Total Population Deaths Percentage
Multnomah 812,855 1,069 0.13%
Marion 347,818 662 0.19%
Clackamas 418,187 538 0.13%
Washington 601,592 515 0.09%
Jackson 220,944 492 0.22%

Highest percentage of Deaths

Highest Percent of Deaths
Oregon Counties as of 2022-03-07
County Total Population Deaths Percentage
Harney 7,393 37 0.50%
Josephine 87,487 314 0.36%
Jefferson 24,658 84 0.34%
Douglas 110,980 372 0.34%
Malheur 30,571 101 0.33%